Hi, I am using fractional regression, fracreg in Stata, to analyze how city's high school student graduation rate ( a percentage) responds to the city's characteristics. My question is whether I can rank my dependent variable for the concern of the potential noise in the measure of the dependent variable. The reason I do this is that if I use the continuous raw values of the dependent variable, the Pseudo R Square is really low. Look forward to your suggestion. Thanks!
0 Response to For fractional regression, it is appropriate to rank the dependent variable for the concern that the measure is noisy?
Post a Comment